The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the Membrane CLI globally using the command npm install -g @membranehq/cli@latest. This is a vendor-provided tool necessary for the skill's operations.\n- [COMMAND_EXECUTION]: Various shell commands are executed through the membrane CLI to manage user authentication, connection state, and action discovery.\n- [REMOTE_CODE_EXECUTION]: The skill can dynamically create and run actions on the Membrane platform via membrane action create and membrane action run. These actions are built and executed remotely based on natural language descriptions provided to the vendor's service.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes data from external RD Station records.\n
Ingestion points: Data enters the agent context through the output of membrane action run and membrane action list when interacting with RD Station records.\n
Boundary markers: The skill instructions do not specify any delimiters or instructions to ignore potential commands embedded within the retrieved record data.\n
Capability inventory: The agent has the capability to execute shell commands using the membrane CLI and create new actions.\n
Sanitization: There is no evidence of sanitization or validation performed on the external data before it is processed by the agent.

rd-station